Predictive value of isolated DLCO reduction
in systemic sclerosis patients without
cardio-pulmonary involvement at baseline
M. Colaci1, D. Giuggioli1, M. Sebastiani1, A. Manfredi1,
F. Lumetti1, F. Luppi2, S. Cerri2, C. Ferri1
1Rheumatology Unit, University of Modena and Reggio Emilia, Medical School, University Hospital of Modena, Italy; 2Respiratory Disease Unit, University of Modena and Reggio Emilia, Medical School, University Hospital of Modena, Italy
summary
Impaired diffusing capacity of the lung for carbon monoxide (DLCO) was frequently observed in systemic sclerosis (SSc) patients, generally related to the presence of interstitial lung disease (ILD) and/or pulmonary arterial hypertension (PAH). However, in clinical practice abnormally low DLCO values may be found also in the absence of these SSc complications. The objective was to investigate the prospective clinical relevance of isolated DLCO reduction at baseline in SSc patients.
Ninety-seven SSc female patients (age at the diagnosis: 51.3±14.5 years; disease duration: 10.4±6.6 years; limited/diffuse skin subsets: 92/5), without any clinical, radiological (high resolution computed tomography), and echocardiographic manifestations of ILD or PAH at baseline, nor other lung or heart diseases able to affect DLCO, were recruited at our Rheumatology Centre.
Patients with DLCO <55% (15 patients; group A) were compared with those with normal DLCO (82 patients; group B), at baseline and at the end of follow-up.
At baseline, patients of group A showed significantly higher percentage of anticentromere autoantibodies com-pared to group B (13/15, 86.6% vs 48/82, 58.5%; p=0.044). More interestingly, at the end of long-lasting clinical follow-up (11.6±6.7 years), pre-capillary PAH (right heart catheterization) solely developed in some patients of group A (3/15, 20% vs 0/82; p=0.003).
In SSc patients, the presence at baseline of isolated, marked DLCO reduction (<55% of predicted) and serum anticentromere autoantibodies might characterize a peculiar SSc subset that may precede the development of PAH. Therefore, careful clinical follow-up of patients with isolated moderate-severe DLCO reduction should be mandatory.
Key words: Scleroderma; systemic sclerosis; DLCO; pulmonary arterial hypertension; lung fibrosis.
Reumatismo, 2015; 67 (4): 149-155
n IntroductIon
S
ystemic sclerosis (SSc) is aconnec-tive tissue disease characterized by excessive production of fibrotic tissue and endothelial dysfunction (1). The latter is responsible of widespread microvascular damage, leading to Raynaud’s phenom-enon, ischemic digital ulcers, pulmonary arterial hypertension (PAH), and/or renal crisis. Indeed, the clinical features of SSc patients may be widely variable, from subclinical vascular abnormalities, until severe life-threatening organ involvement (2). Therefore, long-term follow-up of SSc patients is recommendable, in order
to diagnose SSc vascular complications precociously, with timely therapeutic ap-proach. Namely, PAH is one of the most se-rious complications of SSc, characterized by very poor 5-year prognosis, generally worse than that observed in patients with idiopathic PAH (3). Besides, the availabil-ity of novel treatments for PAH, mainly the anti-endothelin antagonists, has given the hope to counteract this devastating compli-cation; therefore, early diagnosis of SSc-associated PAH may be decisive for the overall disease outcome (4). Transthoracic color-Doppler echocardiography is a non-invasive, reproducible, widely available tool to accurately estimate pulmonary
arte-Corresponding author Michele Colaci Rheumatology Unit,
University of Modena and Reggio Emilia Policlinico di Modena
Via del Pozzo, 71 - 41100 Modena, Italy E-mail: michelecolaci@virgilio.it
Non-commercial
rial pressure and to evaluate right ventricle performance (5). PAH is suspected if the peak velocity of tricuspid regurgitation is >3 meters/second, with consequent sys-tolic pulmonary arterial pressure (PAPs) resulted >35 mmHg (6); in this case, the right heart catheterization is mandatory in order to confirm the vascular alteration and to distinguish between pre-capillary and post-capillary PAH (5, 6). Apart from cardiac ultrasonography, the presence of PAH can be suspected on the basis of the reduction of diffusing capacity of the lung for carbon monoxide (DLCO), mainly in absence of interstitial lung disease (ILD) (7). Recently, an international cross-sec-tional study (DETECT study) proposed an algorithm, in order to identify PAH in SSc patients and to minimise missed diagnoses (8). This tool includes forced vital capac-ity (FVC)%/DLCO% ratio, telangiectasias, anticentromere antibodies, NTproBNP, se-rum urate, and right axis deviation at elec-trocardiogram, as parameters to be consid-ered in order to select patients who should perform echocardiography. In the second step, the referral to right heart catheteriza-tion was decided after considering right atrium area and tricuspid regurgitant jet ve-locity (TV) at echocardiography.
Interestingly, a few SSc patients may pres-ent an apparpres-ently isolated reduction of DLCO, without evidencing ILD at chest high-resolution computed tomography (HRCT) and with normal parameters (i.e., TV) at echocardiographic evaluation. Giv-en the proper execution of the test, with correction for patients’ hemoglobin levels (9), and by excluding other known causes that may reduce DLCO, the clinico-prog-nostic significance of this finding remains unknown. Hypothetically, SSc-related sub-clinical microvascular lung involvement could be suspected, which may potentially evolve into overt PAH. A predictive role of DLCO reduction, at different cut-off levels, for PAH development has been suggested with variable sensitivity and specificity (10); however, these previous studies ana-lyzed the predictive value of DLCO reduc-tion in the setting of patients with one or more cardio-pulmonary parameter
altera-tions. On the other hand, the long-term out-come of SSc patients with isolated DLCO impairment was investigated by Steen et al. (11) in a cohort of 152 subjects recruited from 1972 to 1987. Among these patients with a mean initial DLCO value of 48% of predicted, 11% developed PAH (p<0.01); moreover, the best predictor for isolat-ed pulmonary hypertension was a FVC/ DLCO ratio >1.4.
The aim of the present study is to inves-tigate the clinico-prognostic relevance of isolated DLCO reduction during the fol-low-up of a cohort of SSc patients in ab-sence of other cardio-pulmonary manifes-tations evaluated by means of chest HRCT and echocardiography.
n MATERIALS AND METHODS
Ninety-seven SSc patients [mean age at the diagnosis 51.3±14.5 standard deviation (SD) years, disease duration at the end of follow-up 10.4±6.6 SD years] referring at our University-based Rheumatology
Cen-tre from January 1st 2002 until December
31th 2012 were considered. All patients
were diagnosed by trained rheumatologists with consolidated expertise in SSc; more-over, they fulfilled the 2013 American Col-lege of Rheumatology (ACR)/European League Against Rheumatism (EULAR) Classification Criteria for SSc (12). At baseline, the inclusion criteria to enter into the study were:
1) absence of dyspnea [New York Heart Association (NYHA) III-IV];
2) absence of ILD at chest HRCT; 3) FVC >80%;
4) normal echocardiography, namely with-out increase of the right atrial or ventric-ular areas and peak velocity of tricuspid regurgitation <3 meters/second;
5) absence of chronic obstructive pulmo-nary disease, heart failure or valvular heart diseases.
The patients were divided in two groups according to the DLCO values, consider-ing as cut-off DLCO < 55%, as proposed by Steen et al. (11).
Pulmonary function tests were performed using the spirometer MasterScreen Body
Non-commercial
(Jaeger-Toennies, Hoechberg, Germany); DLCO was measured by a single-breath technique and corrected for hemoglobin. Results were expressed as the percentage of predicted values; values <80% of predicted were considered abnormal, using reference values of the Official Statement of the Eu-ropean Respiratory Society. Chest HRCT scans were obtained using the LightSpeed VCT (GE Healthcare, Buckinghamshire, UK). One-mm sections were acquired at 10-mm intervals from apex to base of lungs, with the patients at full inspiration in a supine position. Additional scans in prone position were eventually recorded to evalu-ate the effect of gravity-dependent opacifi-cation. Echocardiographic measurements were performed with the subjects in partial left decubitus, by digital Vivid 7 ultrasound machine (GE-Vingmed, Horten, Norway).
A variable frequency phased-array trans-ducer (2.5-3.5-4.0 MHz) was used for two-dimensional, M-mode, and Doppler imaging. Left ventricular measures were
obtained by M-modewith 2-D guidance at
conventional levels, according to the
Amer-icanSociety of Echocardiography
recom-mendations. Right ventricle (RV) end-diastolic diameter was measured in apical four-chamber view at middle level. PAPs were estimated using Doppler recordings of tricuspid regurgitation, measuring the peak velocity of the tricuspid regurgita-tion jet. According to the data of the French Nationwide Prospective Multicenter Study, we considered significant the peak velocity that resulted >3 meters/second (6).
During the follow-up, spirometry with DLCO test and echocardiography were routinely performed yearly.
Table I - Clinical features of the whole systemic sclerosis series and 2 groups according to different DlCo values at baseline and at the end of follow-up.
Baseline End of Follow-up
ssc series Group a Group B p
ssc series Group a Group B p No. 97 (DLCO <55%) (DLCO ≥55%) No. 97 (DLCO <55%) (DLCO ≥55%)
N=15 N=82 N=15 N=82
age (years) 51.3±14.5 52.9±10.5 49±13.6 ns 60.1±12.6 64.4±9.8 59.3±13 ns Dis. duration (years) 1 (0-28)* 2 (0-13)1 1 (0-28)* ns 10.4±6.6 11.5±7.1 10.3±6.3 ns
Skin subset (l/D) 93/4 14/1 79/3 ns 92/5 14/1 78/4 ns Digital ulcers° 22 (22.7%) 3 (20%) 19 (23.2%) ns 32 (33%) 6 (40%) 26 (31.7%) ns esophageal involvement 51 (52.6%) 9 (60%) 42 (51.2%) ns 60 (61.8%) 11 (73.3%) 49 (59.7%) ns aCa 61 (62.9%) 13 (86.6%) 48 (58.5%) 0.044 anti-Scl70 28 (28.9%) 2 (13.3%) 26 (31.7%) ns anoa 6 (6.2%) 0 6 (7.3%) ns Heart involvement 0 0 0 ns 4 (4.1%) 2 (13.3%) 2 (2.4%) ns lung fibrosis 0 0 0 ns 14 (14.4%) 4 (26.7%) 9 (11%) ns FVC% 106.7±13.8 106.3±13 106.6±14.1 ns 107.7±17.6 101.8±18 108.1±18.6 ns DlCo% 65.3±16.5 46±6.9 75.2±14.4 <0.001 64.6±15.7 44.5±7 68.2±14 <0.001 FVC/DlCo>1.6 97 (30.9%) 15 (100%) 15 (18.3%) <0.001 58 (59.8%) 15 (100%) 43 (52.4%) <0.001 paH 0 0 0 ns 3 (3.1%) 3 (20%) 0 0.003 Treatment - iloprost# 10 (10.3%) 1 (6.7%) 9 (11%) ns - Bosentan§ 6 0 5 (6.1%) ns
SSc, systemic sclerosis; DlCo, diffusing capacity of the lung for carbon monoxide; Dis., disease; FVC, forced vital capacity; paH, pulmonary arterial hypertension. *Median (range); °at least one episode during the time period preceding the patients’ evaluation; #for at least 5 years before the last patients’ evaluation; §for at least 3 years before the last patients’ evalutation.
Non-commercial
Statistical analysis was performed in or-der to investigate the possible correlations between SSc parameters and DLCO re-duction. Values are given as mean±SD for normally distributed variables, or as me-dian (range) for not normally distributed variables. Group values and proportions were compared by univariate analysis of variance (ANOVA) and Fisher’s exact test, respectively.
n RESULTS
Demographic and clinico-serological fea-tures of 97 SSc patients included in the study are shown in Table 1. Considering the 55% cut-off for DLCO, 15 patients with these DLCO levels were included in group A, while 82 constituted group B (DLCO ≥55%).
All patients were females; except from 5 cases, all individuals presented limited skin subset according to LeRoy’s classification (13). Moreover, 61 had anticentromere, 28 anti-Scl70, and 7 anti-nucleolar autoanti-bodies. According to inclusion criteria, no subject showed cardio-pulmonary manifes-tations at baseline; consistently, mean FVC was 106.7±13.8%. Evaluating possible differences between the 2 subsets, a sig-nificant higher frequency of ACA autoan-tibodies was found in the first group [13/15 (86.6%) vs 48/82 (58.5%); p=0.044]. At the end of the clinical follow-up (11.6±6.7 years), 14/97 patients developed ILD (4 in group A and 9 in group B; p=not significant), represented by mild bi-basilar fibrosis with ground-glass pattern at chest HRCT; indeed, mean FVC% values did not decrease significantly.
Of note, three subjects in group A devel-oped pre-capillary PAH (p=0.003), diag-nosed by right heart catheterization. At the end of follow-up, another patient in group A presented a new increase of PAPs mea-sured by means of cardiac ultrasonogra-phy, but suspected PAH was not confirmed at right heart catheterization. Interestingly, the disease duration of the patients with pre-capillary PAH was >20 years in 2, and 8 years in the third one.
With regards to DLCO values, a significant
deterioration (DLCO reduction >10% of baseline values) was observed in 19 patients of group B during the follow-up; among them, 9 cases presented DLCO values <55% at the end of the study. These subjects pre-sented a disease duration of 14.1±8.1 years; 7/9 were ACA positive and 4 out of them presented digital ulcers; one patient devel-oped a diffuse skin involvement and lung fi-brosis; in the end, the last patient developed a scleroderma renal crisis.
Concordantly to DLCO subsetting, a FVC/ DLCO ratio >1.6 was invariably (100%) observed in patients of group A and in 18.3% of group B, respectively (p<0.001); this statistically significant difference among two groups was also present at the end of follow-up (100% vs 52.4%, respec-tively; p<0.001).
The possible influence of vasoactive treat-ments was also analyzed, particularly some therapies (i.e., iloprost, bosentan) frequently used in SSc patients with PAH. Indeed, even because of the low number of patients treat-ed, no statistical correlations were recorded.
n DISCUSSION
In the present study, the prospective clini-cal significance of markedly low DLCO values, in absence of ILD or PAH at base-line, was investigated. In particular, SSc patients with isolated reduction of DLCO (<55%) at the beginning of the follow-up showed higher prevalence of serum ACA autoantibodies compared to patients with normal DLCO values, as well as signifi-cantly increased incidence of PAH. DLCO values represent a measure of the resistance encountered by gases during alveolar-capillary membrane diffusion. Therefore, every condition affecting the physiological gas exchanges may be esti-mated by means of DLCO test (14). Im-paired DLCO was frequently reported in SSc patients; two main pathological con-ditions may be responsible of this altera-tion, namely ILD and PAH (7, 15). The first one may affect DLCO levels by re-ducing alveolar volume and/or by alveo-lar–capillary membrane thickening (16). This latter may be impaired also by PAH,
Non-commercial
in which vascular remodelling leads to arterial stiffness and vessel wall tighten-ing (17). In our study, we selected SSc patients without any evidence of ILD or PAH at baseline, in order to eliminate the main causes of DLCO impairment. However, a significant group of patients (36/97; 37.1%) presented DLCO values <70%, not explained by other confound-ing factors, such as anaemia, smoke habit, or presence of other lung or heart diseases known to influence the test (14). Isolated DLCO reduction has been reported in SSc, especially in patients with limited skin subset (18), consistently with our findings.
In fact, DLCO impairment has been in-cluded among diagnostic parameters in patients with early SSc (8, 19), also in absence of overt organ involvement. However, it is more difficult to attribute specific clinico-prognostic significance to this alteration, including possible early therapeutic attempts, in subjects generally asymptomatic or with very mild clinical symptoms.
On the other hand, the presence of mod-erate-severe reduction of DLCO, accord-ingly to the cut-off level here employed (<55%), should induce at least patient’s careful evaluation and monitoring. It was well established that DLCO reduction is one of the best predictor of PAH devel-opment, particularly in presence of FVC within normal range; namely, a FVC/ DLCO ratio of 1.6-1.8 helps to identify an increased likelihood of having PAH (11, 20). This issue has been focused by a multicenter cohort study for PAH assess-ment and recognition of outcomes (21); in particular, SSc patients with DLCO <55% of predicted or FVC/DLCO >1.6, and PAPs >40 mmHg, were considered at-risk for PAH (22). However, consid-ering patients with complicating PAH versus those without, low DLCO alone at baseline was not able to distinguish between two groups, even if it remains an important predictor of this complica-tion. Moreover, severely reduced DLCO (<39%) at the time of PAH diagnosis seems to be a worse prognostic factor in
SSc patients. Consistently, Arunthari et al. (23) did not found significant correla-tions between right heart catheterization parameters and DLCO values in a cohort of 138 individuals, even when corrected for spirometric volumes.
Accordingly, we did not found a signifi-cant correlation between DLCO levels at baseline and PAH development consider-ing a DLCO reduction <70% as cut-off (data not shown); otherwise, focusing on the patients’ subset with lower DLCO lev-els (<55%) a statistically significant cor-relation was observed.
In the DETECT study (8), evaluating 466 SSc patients at increased risk of PAH, a value of DLCO <60% was an inclusion cri-terion; the 19% of these patients presented PAH, consistently with the findings of our present series (3/15, 20%). Nonetheless, other parameters have been encompassed in the DETECT algorithm, since also echo-cardiography alone showed to be not so accurate for PAH identification (8). In this respect, considering our study, it cannot be excluded the possibility to have overlooked initial/mild PAH patients on the basis of normal findings obtained at the baseline echocardiography; however, this occur-rence has been eventually offset by the long-lasting follow-up, including yearly echocardiographic evaluations.
With regards to other clinical SSc features, a trend to higher frequency of digital ulcers in patients with reduced DLCO was found at the end of follow-up. This findings, along with the significant presence of ACA autoantibodies and the pretension to PAH development, might define a peculiar SSc subset with significant vascular involve-ment. As regards skin subset, the low num-ber of patients with diffuse skin involve-ment did not permit a conclusive statistical analysis.
In our patients’ series, disease duration seems to be quite relevant in the develop-ment of PAH in two patients. In fact, 2/3 subjects presented this complication after >20 years of follow-up; in particular, PAH appeared after 21 and 22 years from SSc diagnosis, respectively. This finding was not constantly reported in the few studies
Non-commercial
previously published; Hsu et al. (22) in the PHAROS study observed the devel-opment of PAH after 10.1±8.4 years from SSc onset. On the contrary, in a multicenter French study (24), PAH developed at any time after SSc diagnosis; in addition, pa-tients with early-onset PAH were older at the time of SSc diagnosis compared to those without (25).
The possible usefulness of anti-endothelin receptor antagonists in preventing PAH re-mains still to be verified.
A limit of our study is the relatively low number of patients taken into account with respect to the low incidence of PAH, one of the main SSc complication potentially cor-related to reduced DLCO (26). However, this single Centre study firstly focused on particularly homogeneous subset of 15 SSc patients, with isolated marked reduction of DLCO in the absence of ILD or PAH, thor-oughly evaluated by means of wide clinical and instrumental assessment; in addition, the patients were compared to a large con-trol group, at baseline and after a long-term clinical follow-up for the detection of late SSc complications.
Another limit of the study is that not all subjects underwent a right-heart-cathe-terization during the follow-up period, because of the lack of suspect of PAH at echocardiography; indeed, SSc patients who developed PAH could be at least un-derestimated, thus our findings might be further reinforced.
n CONCLUSIONS
In conclusion, it could be hypothesized that the presence of clear-cut isolated DLCO impairment, serum ACA autoan-tibodies, and the propensity to develop PAH or digital ulcers might characterize a peculiar SSc subset; therefore, careful clinical follow-up of patients with iso-lated DLCO reduction (<55%) should be mandatory in order to early detect the ap-pearance of PAH, one of the most harm-ful SSc complication.
Conflict of interests: the authors declare to
have no conflict of interests.
n REfERENCES
1. Hudson M, Fritzler MJ. Diagnostic criteria of systemic sclerosis. J Autoimm. 2014; 48-49: 38-41.
2. Shah AA, Wigley FM. My approach to the treatment of scleroderma. Mayo Clin Proc. 2013; 88: 377-93.
3. Fisher MR, Mathai SC, Champion HC, et al. Clinical differences between idiopathic and scleroderma-related pulmonary hypertension. Arthritis Rheum. 2006; 54: 3043-50.
4. Hachulla E, Denton CP. Early intervention in pulmonary arterial hypertension associated with systemic sclerosis: an essential compo-nent of disease management. Eur Respir Rev. 2010; 19: 314-20.
5. Hoeper MM, Bogaard HJ, Condiffe R, et al. Definitions and diagnosis of pulmonary hyper-tension. J Am Coll Cardiol. 2013; 62: D42-50. 6. Hachulla E, Gressin V, Guillevin L, et al. Ear-ly detection of pulmonary arterial hyperten-sion in systemic sclerosis. A French Nation-wide Prospective Multicenter Study. Arthritis Rheum. 2005; 52: 3792-800.
7. Sivova N, Launay D, Wémeau-Stervinou L, et al. Relevance of partitioning DLCO to detect pulmonary hypertension in systemic sclerosis. PLoS One. 2013; 8: e78001.
8. Coghlan JG, Denton CP, Grünig E, et al. Evidence-based detection of pulmonary ar-terial hypertension in systemic sclerosis: the DETECT study. Ann Rheum Dis. 2014; 73: 1340-9.
9. Hughes JMB, Pride NB. Examination of the carbon monoxide diffusing capacity (DLco) in relation to its Kco and Va components. Am J Respir Crit Care Res. 2012; 186: 132-9. 10. Gladue H, Altorok N, Townsend W, et al.
Screening and diagnostic modalities for con-nective tissue disease-associated pulmonary arterial hypertension: a systematic review. Se-min Arthritis Rheum. 2014; 43: 536-41. 11. Steen VD, Graham G, Conte C, et al. Isolated
diffusing capacity reduction in systemic scle-rosis. Arthritis Rheum. 1992; 35: 765-70. 12. van den Hoogen F, Khanna D, Fransen J, et al.
2013 classification criteria for systemic scle-rosis: an American College of Rheumatology/ European League Against Rheumatism col-laborative initiative. Ann Rheum Dis. 2013; 72: 1747-55.
13. LeRoy EC, Black C, Fleischmajer R, et al. Scleroderma (systemic sclerosis): classifica-tion, subsets and pathogenesis. J Rheumatol. 1988; 15: 202-5.
14. Cotton DJ, Graham BL. Single-breath carbon monoxide diffusing capacity or transfer factor. In: Qutayba H, Joanne S, James M, eds. Physi-ologic basis of respiratory disease. Hamilton: BC Decker Inc. 2005; 659-9.
15. van der Lee I, Zanen P, Grutter JC, et al. Dif-fusing capacity for nitric oxide and carbon
Non-commercial
monoxide in patients with diffuse parencmal lung disease and pulmonary arterial hy-pertension. Chest. 2006; 129: 378-83. 16. Hughes JM. The single breath transfer factor
(TLCO) and the transfer coefficient (KCO): a window onto the pulmonary microcirculation. Clin Physiol Funct Imaging. 2003; 23: 63-71. 17. Chandra S, Shah SJ, Thenappan T, et al. Car-bon monoxide diffusing capacity and mortal-ity in pulmonary arterial hypertension. J Heart Lung Transplant. 2010; 29: 181-7.
18. Steen V, Medsger TA Jr. Predictors of isolated pulmonary hypertension in patients with sys-temic sclerosis and limited cutaneous involve-ment. Arthritis Rheum. 2003; 48: 516-22. 19. Trad S, Huong du LT, Frances C, et al.
Im-paired carbon monoxide diffusing capacity as a marker of limited systemic sclerosis. Eur J Int Med. 2011; 22: e80-6.
20. Denton CP, Calies JB, Phillips GD, et al. Comparison of doppler echocardiography and right heart catheterization to assess pulmonary hypertension in systemic sclerosis. Br J Rheu-matol. 1997; 36: 239-43.
21. Hinchcliff M, Fischer A, Schiopu E, Steen V. Pulmonary hypertension assessment and rec-ognition of outcomes in scleroderma (PHAR-OS): baseline characteristics and description
of study population. J Rheumatol. 2011; 38: 2172-9.
22. Hsu VM, Chung L, Hummers LK, et al. De-velopment of pulmonary hypertension in a high-risk population with systemic sclerosis in the Pulmonary Hypertension Assessment and Recognition of Outcomes in Scleroder-ma (PHAROS) cohort study. Semin Arthritis Rheum. 2014; 44: 55-62.
23. Arunthari V, Burger CD, Lee AS. Correlation of pulmonary function variables with hemo-dynamic measurements in patients with pul-monary arterial hypertension. Clin Respir J. 2011; 5: 35-43.
24. Hachulla E, Launay D, Mouthon L, et al. Is pulmonary arterial hypertension really a late complication of systemic sclerosis? Chest. 2009; 136: 1211-9.
25. Medsger TA Jr. Natural history of systemic sclerosis and the assessment of disease activ-ity, severactiv-ity, functional status, and psychologic well-being. Rheum Dis Clin North Am. 2003; 29: 255-73.
26. Iudici M, Codullo V, Giuggioli D, et al. Pul-monary hypertension in systemic sclerosis: prevalence, incidence and predictive factors in a large multicentric Italian cohort. Clin Exp Rheumatol. 2013; 31: 31-6.
